Merge branch 'akpm' (Andrew's incoming - part two)

Says Andrew:

 "60 patches.  That's good enough for -rc1 I guess.  I have quite a lot
  of detritus to be rechecked, work through maintainers, etc.

 - most of the remains of MM
 - rtc
 - various misc
 - cgroups
 - memcg
 - cpusets
 - procfs
 - ipc
 - rapidio
 - sysctl
 - pps
 - w1
 - drivers/misc
 - aio"

* akpm: (60 commits)
  memcg: replace ss->id_lock with a rwlock
  aio: allocate kiocbs in batches
  drivers/misc/vmw_balloon.c: fix typo in code comment
  drivers/misc/vmw_balloon.c: determine page allocation flag can_sleep outside loop
  w1: disable irqs in critical section
  drivers/w1/w1_int.c: multiple masters used same init_name
  drivers/power/ds2780_battery.c: fix deadlock upon insertion and removal
  drivers/power/ds2780_battery.c: add a nolock function to w1 interface
  drivers/power/ds2780_battery.c: create central point for calling w1 interface
  w1: ds2760 and ds2780, use ida for id and ida_simple_get() to get it
  pps gpio client: add missing dependency
  pps: new client driver using GPIO
  pps: default echo function
  include/linux/dma-mapping.h: add dma_zalloc_coherent()
  sysctl: make CONFIG_SYSCTL_SYSCALL default to n
  sysctl: add support for poll()
  RapidIO: documentation update
  drivers/net/rionet.c: fix ethernet address macros for LE platforms
  RapidIO: fix potential null deref in rio_setup_device()
  RapidIO: add mport driver for Tsi721 bridge
  ...

1 files changed, 26 insertions, 11 deletions

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 860ec211ddd6..4298abaae153 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -990,7 +990,7 @@ struct page *follow_trans_huge_pmd(struct mm_struct *mm,
 	page += (addr & ~HPAGE_PMD_MASK) >> PAGE_SHIFT;
 	VM_BUG_ON(!PageCompound(page));
 	if (flags & FOLL_GET)
-		get_page(page);
+		get_page_foll(page);
 
 out:
 	return page;
@@ -1202,6 +1202,7 @@ static void __split_huge_page_refcount(struct page *page)
 	unsigned long head_index = page->index;
 	struct zone *zone = page_zone(page);
 	int zonestat;
+	int tail_count = 0;
 
 	/* prevent PageLRU to go away from under us, and freeze lru stats */
 	spin_lock_irq(&zone->lru_lock);
@@ -1210,11 +1211,27 @@ static void __split_huge_page_refcount(struct page *page)
 	for (i = 1; i < HPAGE_PMD_NR; i++) {
 		struct page *page_tail = page + i;
 
-		/* tail_page->_count cannot change */
-		atomic_sub(atomic_read(&page_tail->_count), &page->_count);
-		BUG_ON(page_count(page) <= 0);
-		atomic_add(page_mapcount(page) + 1, &page_tail->_count);
-		BUG_ON(atomic_read(&page_tail->_count) <= 0);
+		/* tail_page->_mapcount cannot change */
+		BUG_ON(page_mapcount(page_tail) < 0);
+		tail_count += page_mapcount(page_tail);
+		/* check for overflow */
+		BUG_ON(tail_count < 0);
+		BUG_ON(atomic_read(&page_tail->_count) != 0);
+		/*
+		 * tail_page->_count is zero and not changing from
+		 * under us. But get_page_unless_zero() may be running
+		 * from under us on the tail_page. If we used
+		 * atomic_set() below instead of atomic_add(), we
+		 * would then run atomic_set() concurrently with
+		 * get_page_unless_zero(), and atomic_set() is
+		 * implemented in C not using locked ops. spin_unlock
+		 * on x86 sometime uses locked ops because of PPro
+		 * errata 66, 92, so unless somebody can guarantee
+		 * atomic_set() here would be safe on all archs (and
+		 * not only on x86), it's safer to use atomic_add().
+		 */
+		atomic_add(page_mapcount(page) + page_mapcount(page_tail) + 1,
+			   &page_tail->_count);
 
 		/* after clearing PageTail the gup refcount can be released */
 		smp_mb();
@@ -1232,10 +1249,7 @@ static void __split_huge_page_refcount(struct page *page)
 				      (1L << PG_uptodate)));
 		page_tail->flags |= (1L << PG_dirty);
 
-		/*
-		 * 1) clear PageTail before overwriting first_page
-		 * 2) clear PageTail before clearing PageHead for VM_BUG_ON
-		 */
+		/* clear PageTail before overwriting first_page */
 		smp_wmb();
 
 		/*
@@ -1252,7 +1266,6 @@ static void __split_huge_page_refcount(struct page *page)
 		 * status is achieved setting a reserved bit in the
 		 * pmd, not by clearing the present bit.
 		*/
-		BUG_ON(page_mapcount(page_tail));
 		page_tail->_mapcount = page->_mapcount;
 
 		BUG_ON(page_tail->mapping);
@@ -1269,6 +1282,8 @@ static void __split_huge_page_refcount(struct page *page)
 
 		lru_add_page_tail(zone, page, page_tail);
 	}
+	atomic_sub(tail_count, &page->_count);
+	BUG_ON(atomic_read(&page->_count) <= 0);
 
 	__dec_zone_page_state(page, NR_ANON_TRANSPARENT_HUGEPAGES);
 	__mod_zone_page_state(zone, NR_ANON_PAGES, HPAGE_PMD_NR);